BenchAtlas

Rankings / OpenAI

GPT 4o (2024-08-06)

released 2024-08-06

BenchAtlas Index

as of 2026-07-14
30.8
high effort
rank #349
6 families · 3 categories · medium
27.7
base configuration
rank #373
5 families · 3 categories · medium

Benchmark evidence

42 results
coding
LiveCodeBench v5older version
problems=279 · window_start=2024-08-01
30.0 %independentLiveCodeBench Leaderboard
LiveCodeBench v6
problems=454 · window_start=2024-08-01
29.5 %independentLiveCodeBench Leaderboard
external indices
ECI129.2 pointsindependentEpoch AI Benchmarking Hub
human preference
Coding (style control)older version
arena=text · category=coding · style_control=true
1360.5
1352.6–1368.4
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1348.2
1342.8–1353.5
communityLMArena Leaderboard Dataset
Industry Life And Physical And Social Science (style control)older version
arena=text · category=industry_life_and_physical_and_social_science · style_control=true
1348.1
1340.1–1356.2
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1347.5
1335.1–1360.0
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1347.2
1340.4–1353.9
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1341.8
1327.9–1355.6
communityLMArena Leaderboard Dataset
Industry Writing And Literature And Language (style control)older version
arena=text · category=industry_writing_and_literature_and_language · style_control=true
1340.2
1333.3–1347.1
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1339.3
1331.5–1347.0
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1338.0
1329.5–1346.4
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1335.1
1330.9–1339.3
communityLMArena Leaderboard Dataset
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1329.0
1318.4–1339.6
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1326.7
1320.3–1333.2
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1326.5
1318.1–1335.1
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1326.4
1318.5–1334.3
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
1325.7
1298.7–1352.8
communityLMArena Leaderboard Dataset
Industry Entertainment And Sports And Media (style control)older version
arena=text · category=industry_entertainment_and_sports_and_media · style_control=true
1324.7
1316.4–1332.9
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1323.2
1314.4–1332.0
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1318.2
1312.4–1324.0
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1315.4
1309.9–1321.0
communityLMArena Leaderboard Dataset
German (style control)older version
arena=text · category=german · style_control=true
1313.1
1295.8–1330.4
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1312.8
1304.5–1321.2
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1309.6
1296.8–1322.3
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1308.6
1300.6–1316.6
communityLMArena Leaderboard Dataset
Industry Business And Management And Financial Operations (style control)older version
arena=text · category=industry_business_and_management_and_financial_operations · style_control=true
1302.5
1293.4–1311.7
communityLMArena Leaderboard Dataset
Japanese (style control)older version
arena=text · category=japanese · style_control=true
1291.8
1273.1–1310.5
communityLMArena Leaderboard Dataset
Korean (style control)older version
arena=text · category=korean · style_control=true
1288.4
1263.9–1312.9
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1285.8
1279.3–1292.3
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond49.2 %independentEpoch AI Benchmarking Hub
2025-01-27
MMLU-Pro
implementation=vals-ai
high effort74.1 %independentVals AI
multimodal
MMMU
implementation=vals-ai
high effort64.0 %independentVals AI
professional
CorpFinhigh effort39.4 %independentVals AI
LegalBenchhigh effort80.1 %independentVals AI
MedQAhigh effort88.2 %independentVals AI
TaxEvalhigh effort71.1 %independentVals AI
reasoning math
AIME
implementation=vals-ai
high effort14.0 %independentVals AI
FrontierMath0.3 %independentEpoch AI Benchmarking Hub
2025-03-07
MATH-500
implementation=vals-ai
high effort75.2 %independentVals AI
MATH Level 553.3 %independentEpoch AI Benchmarking Hub
2025-01-27
OTIS Mock AIME 2024–20256.4 %independentEpoch AI Benchmarking Hub
2025-02-25

Agent + model results

systems, not bare-model scores
agent + model OpenCSG Starship Agentic Coder + GPT 4o (2024-08-06)SWE-bench Lite39.7 %unverifiedSWE-bench Leaderboard
agent + model MASAI + GPT 4o (2024-08-06)SWE-bench Verified32.6 %unverifiedSWE-bench Leaderboard
agent + model GUIRepair + GPT 4o (2024-08-06)SWE-bench Multimodal30.4 %communitySWE-bench Leaderboard
agent + model Agentless + GPT 4o (2024-08-06)SWE-bench Lite29.7 %unverifiedSWE-bench Leaderboard
agent + model reproducedRG + GPT 4o (2024-08-06)SWE-bench Lite28.0 %unverifiedSWE-bench Leaderboard
agent + model EPAM AI/Run Developer Agent + GPT 4o (2024-08-06)SWE-bench Verified27.0 %unverifiedSWE-bench Leaderboard
agent + model HyperAgent + GPT 4o (2024-08-06)SWE-bench Lite25.3 %unverifiedSWE-bench Leaderboard
agent + model Bytedance AutoSE (based on SWE-Agent) + GPT 4o (2024-08-06)SWE-bench Lite21.7 %unverifiedSWE-bench Leaderboard
agent + model RAG + GPT 4o (2024-08-06)SWE-bench Multimodal6.0 %communitySWE-bench Leaderboard
agent + model Agentless + GPT 4o (2024-08-06)SWE-bench Multimodal3.1 %communitySWE-bench Leaderboard

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.

Related variants

GPT 4.5 (2025-02-27)GPT 4o (2024-11-20)GPT 4.1 (2025-04-14)GPT 5 (2025-08-07)GPT 5.1 (2025-11-13)GPT 5.2 (2025-12-11)GPT 5.5 (2026-04-22)GPT 5.4 (2026-03-05)